Apoptosis occurs in the placenta throughout gestation, with a greater frequency near term in comparison to the first trimester. The Fas/FasL system represents one of the main apoptotic pathways controlling placental apoptosis. Although first trimester trophoblast cells express both Fas and FasL, they are resistant to Fas‐induced apoptosis. Therefore, trophoblast resistance to Fas‐mediated apoptosis may be due to the inhibition of the pathway downstream of Fas stimulation. Expression levels of X‐linked inhibitor of apoptosis (XIAP) were recently shown to decrease in third trimester placentas, correlating with an increase in placental apoptosis. As a potent caspase inhibitor, XIAP prevents the activation of caspase‐9 through its BIR3 domain and caspase‐3 activation via the linker‐BIR2 domain. In the present study, high levels of the active form of XIAP were detected in first trimester trophoblast cells, whereas term placental tissue samples predominantly expressed the inactive form of XIAP. Using a XIAP inhibitor, phenoxodiol, we demonstrate that XIAP inactivation sensitizes trophoblast cells to Fas stimulation, as evidenced by the anti‐Fas mAb‐induced decrease in trophoblast cell viability and increase in caspase‐8, caspase‐9 and caspase‐3 activation. This suggests a functional role for XIAP in the regulation of the Fas apoptotic cascade in trophoblast cells during pregnancy.
Apoptosis, or programmed cell death, is a natural mechanism by which the body eliminates unnecessary or potentially dangerous cells in order to maintain normal tissue function. During implantation, apoptosis is important for the appropriate tissue remodelling of the maternal decidua and invasion of the developing embryo (Uckan et al., 1997; Galan et al., 2000). Apoptosis has also been observed in the placenta throughout gestation; however, a higher frequency occurs in third trimester villi compared with first trimester placentas, suggesting that placental apoptosis is a process that occurs during normal pregnancy (Smith et al., 1997b, 2000). Pregnancies complicated by pre‐eclampsia or intrauterine growth restriction (IUGR) are often associated with insufficient trophoblast invasion, which may be a result of increased trophoblast apoptosis (Smith et al., 1997a; Allaire et al., 2000; Crocker et al., 2003). This suggests that the regulation of trophoblast apoptosis is essential for the normal physiology of pregnancy.
The Fas/FasL system is one of the main apoptotic pathways controlling trophoblast apoptosis. Along with several others, we have shown that villous (Runic et al., 1996; Mor et al., 1998; Payne et al., 1999; Balkundi et al., 2000; Gruslin et al., 2001) and extravillous (Uckan et al., 1997; Kauma et al., 1999; Hammer and Dohr, 2000) trophoblast express both Fas (CD95) and Fas Ligand (FasL; CD95L). Fas and FasL are type I and type II transmembrane proteins of the tumour necrosis factor (TNF) receptor superfamily, respectively (Wallach et al., 1999). The interaction between Fas and FasL or the binding of an agonistic anti‐Fas monoclonal antibody (mAb) to the extracellular region of Fas results in the activation of the Fas receptor (Yonehara et al., 1989). Consequently, the Fas‐associated death domain (FADD) binds to the cytoplasmic tail of Fas and recruits other cellular proteins, forming the death‐inducing signalling complex (DISC) (Scaffidi et al., 1999), the point at which the TNF death receptor pathways converge. Once assembled, procaspase‐8 binds to the DISC and is activated by a series of cleavage steps (Muzio et al., 1996). Upon activation, caspase‐8 ‘initiates’ apoptosis by activating effector caspases either directly or indirectly by feeding the death signal to the mitochondria, which results in caspase‐9 activation. In turn, caspase‐9 activates ‘effector’ caspase‐3 and caspase‐7, the point at which the mitochondrial and death receptor pathways overlap (Hirata et al., 1998). Each step of the pathway is tightly controlled by intracellular inhibitors, which prevent further propagation of the Fas death signal either at the ‘initiator’ or ‘effector’ level.
By precluding caspase‐8 recruitment to the DISC, flice‐like inhibitory protein (FLIP) inhibits apoptosis only triggered by death receptors (Irmler et al., 1997). Similar to FLIP, anti‐apoptotic Bcl‐2 family members prevent the activation of the ‘initiator’ caspase‐9, by inhibiting signals initiated by or directed to the mitochondria (Adams and Cory, 1998; Gross et al., 1999). Unlike FLIP and the anti‐apoptotic members of the Bcl‐2 family, inhibitors of apoptosis (IAP) are unique in that they are capable of inhibiting both the mitochondrial and death receptor‐mediated pathways. To date, eight human IAP have been identified, but X‐linked inhibitor of apoptosis (XIAP) appears to be the most potent and versatile member of the family and is expressed in a variety of tissues (Liston et al., 1996). XIAP contains a single C‐terminal RING (really interesting new gene) domain and three tandem baculoviral IAP repeat (BIR) domains, which have been shown to differentially inhibit initiator and effector caspases (Deveraux et al., 1997). While the BIR1–BIR2 linker, together with the BIR2 domain of XIAP, has been shown to prevent caspase‐3 activation (Takahashi et al., 1998), the caspase‐9 inhibitory activity of XIAP has been localized to the BIR3–RING domain (Sun et al., 2000). Upon Fas stimulation, XIAP was previously shown to be cleaved into two distinct fragments, an N‐terminal fragment containing BIR1‐2 and a second fragment containing BIR3–RING and that this effect was caspase dependent. The BIR1–2 fragment has diminished ability to inhibit caspase‐3 and may be susceptible to further caspase‐mediated degradation. In contrast, the BIR3–RING fragment appears to be more stable and retains the ability to inhibit caspase‐9, but is unable to suppress Fas‐induced apoptosis (Deveraux et al., 1999).
First trimester trophoblast cells express both Fas and FasL, but do not undergo Fas‐mediated apoptosis upon Fas stimulation under normal conditions (Payne et al., 1999; Aschkenazi et al., 2002). Therefore, the expression of Fas does not necessarily correlate with susceptibility to Fas‐induced apoptosis. It has been hypothesized that trophoblast resistance to Fas‐mediated apoptosis may be due to inhibition of the pathway downstream of Fas stimulation to avoid killing by FasL expressed by either the same or neighbouring trophoblast cell (Payne et al., 1999; Aschkenazi et al., 2002). XIAP immunoreactivity was recently identified in the trophoblast layer of first trimester placentas, but not at term, suggesting an anti‐apoptotic role for XIAP in first trimester trophoblast cells (Gruslin et al., 2001). The aim of the present study was to determine if XIAP plays a functional role in the regulation of the apoptotic cascade in trophoblast cells. Using a XIAP inhibitor, phenoxodiol (Kamsteeg et al., 2003), we demonstrate that XIAP inactivation sensitizes trophoblast cells to Fas‐mediated apoptosis. Our findings suggest that XIAP protects trophoblast cells from Fas‐induced apoptosis, which may provide insight into both normal and abnormal placental development.
Materials and methods
Ten term placentas (38–40 weeks gestational age) were obtained from the fetal side of clinically normal pregnancies following vaginal delivery. Ten first trimester placentas between 6 and 12 weeks of gestation were obtained from the fetal side of normal pregnancies, voluntarily terminated for reasons unrelated to the present study. A signed, written consent form was obtained from each patient. The use of placental tissue specimens and consent forms was approved by the Yale University Human Investigation Committee. Tissue specimens were collected in cold, sterile phosphate‐buffered saline (PBS) and immediately transported to the laboratory for cell culture preparation.
Chemicals and antibodies
Phenoxodiol, a synthetic derivative of genestein and a specific inhibitor of XIAP, was obtained from Novogen (USA) (Kamsteeg et al., 2003). The agonistic anti‐human Fas monoclonal antibody (mAb) (clone E0S9.1) was obtained from BD PharMingen (USA). Camptothecin was purchased from Sigma (USA). The mouse anti‐XIAP mAb (clone 28; 1:1000) was purchased from BD Transduction Labs (USA), the mouse anti‐caspase‐8 mAb (clone 1‐3; 1:1000) from Oncogene Research Products (USA), the mouse anti‐caspase‐9 mAb (clone LAP6; 1:2000) from R&D Systems, Inc. (USA) and the rabbit anti‐caspase‐3 polyclonal antibodies, which recognize either all forms (H‐277; 1:500) or the active form of caspase‐3 (9661; 1:5000), were obtained from Santa Cruz Biotechnology, Inc. (USA) and Cell Signaling Technology, Inc. (USA), respectively. The rabbit polyclonal antibody for β‐actin (A2066; 1:10 000) was obtained from Sigma. Primary antibody signals were detected using either a peroxidase‐conjugated horse anti‐mouse, or a peroxidase‐conjugated goat anti‐rabbit secondary antibody from Vector Laboratories (USA).
The first trimester human cytotrophoblast cell line, 3A, which was transformed by SV40 ts30 (Chou, 1978), was purchased from ATCC (USA). The SVneo transformed first trimester human extravillous trophoblast cell line, HTR8 (Graham et al., 1993), hereafter referred to as H8, was a gift from Dr Charles Graham (Queen’s University, Kingston, ON, Canada). Both trophoblast cell lines and Jurkat cells, a human T cell leukaemia line (ATCC), were cultured in Roswell Park Memorial Institute 1640 (Gibco, USA) supplemented with 10% fetal bovine serum (Hyclone, USA), 10 mmol/l HEPES, 0.1 mmol/l minimum essential medium (MEM) non‐essential amino acids, 1 mmol/l sodium pyruvate and 100 IU/ml penicillin/streptomycin (Gibco) and maintained at 37°C in 5% CO2.
First trimester primary trophoblast cell isolation
Primary trophoblast cells were isolated from first trimester placentas according to Loke et al. (1989) with a few modifications (Verwer, 1999). In brief, placentas were washed with cold Hanks’ balanced salt solution (HBSS; Gibco) to remove excess blood. Cells were removed from the membranes by scraping and transferred to trypsin–EDTA (Gibco) digestion buffer and incubated at 37°C for 10 min with shaking. An equal volume of DMEM medium (Gibco) containing 10% fetal bovine serum was added to inactivate the trypsin. This mixture was vortexed for 20 s, allowed to sediment and the supernatant was collected. The two previous steps were repeated twice and the collected supernatant was centrifuged at 300 g for 10 min. Contaminating red blood cells were removed by resuspending the cellular pellet in HBSS, layering this suspension over the same volume of lymphocyte separation media (ICN Biomedicals, Inc., USA) and centrifuging the gradient at 500 g for 25 min. The interface, containing the trophoblast cells, was removed by transfer pipette and cultured at 37°C in 5% in DMEM supplemented with 10% human serum (Gemini Bio‐Products, USA). The purity of the trophoblast cell preparation was >98% as determined by cytokeratin‐7 immunostaining (Dako, USA).
Treatment with the anti‐Fas mAb, phenoxodiol and camptothecin
To evaluate the sensitivity of first trimester trophoblast cells to Fas, the cell lines and primary cultures were treated with 500 ng/ml of the anti‐Fas mAb for 24 h. The effect of phenoxodiol treatment on trophoblast cells was determined by treating the cell lines and primary cultures with 10 µg/ml of phenoxodiol for 24 h. In order to assess whether phenoxodiol treatment sensitized trophoblast cells to Fas, the cell lines and primary cultures were pre‐treated with 1 µg/ml of phenoxodiol for a period of 2 or 18 h (depending on the experiment), after which the cells were treated for an additional 24 h with 1 µg/ml phenoxodiol alone or the combination of phenoxodiol (1 µg/ml) and increasing concentrations of the anti‐Fas mAb (125, 250, 500 ng/ml). As a positive control, the trophoblast cell lines and primary cultures were treated with 4 µmol/l camptothecin for 24 h.
Cell viability assay
Cell viability was evaluated using the CellTiter 96 assay according to the manufacturer’s instructions (Promega, USA) (Neale et al., 2003). Briefly, 1 × 104 cells were plated in triplicate wells in a 100 µl volume per well in a 96‐well microtitre plate (BD Biosciences). Trophoblast cells (cell lines and primary cultures) were grown to 70% confluence, at which stage the medium was replaced with reduced serum phenol‐depleted Opti‐MEM (Gibco) and the cells were cultured for an additional 24 h prior to treatment. Following treatment, 20 µl of the Cell Titer 96 Aqueous One Solution was added to each well and the plate was incubated at 37°C for 1–4 h. Optical densities of the samples were measured at 490 nm using an automatic microplate reader (Model 550; Bio‐Rad, USA). The values of the treated cells were compared with the values generated from the untreated control and reported as percentage viability.
Western blot analysis
Trophoblast cells (5 × 105; cell lines and primary cultures) were plated in 35 mm2 Petri dishes (BD Biosciences) and grown to 70% confluence for treatment. Following treatment, cells were lysed in 1% Nonidet P‐40 and 0.1% sodium dodecyl sulphate (SDS) in the presence of 0.2 mg/ml phenylmethylsulphonyl fluoride and a protease inhibitor cocktail (Roche Applied Science, USA). Placental tissue samples were lysed in 1 ml lysis buffer per 1 g of material following homogenization. Protein concentrations were calculated by BCA assay (Pierce Biotechnology, USA). Total cellular protein (20 µg) was loaded per lane and separated under reducing conditions by SDS–polyacrylamide gel electrophoresis using 12% polyacrylamide gels and transferred to PVDF membranes (NEN Life Sciences, USA) as previously described (Bechmann et al., 1999). The membranes were stained with Ponceau Red to ensure efficient transfer and equal loading of proteins. To inhibit non‐specific binding, membranes were blocked with 5% powdered milk in PBS/0.05% Tween‐20 (PBS‐T) prior to immunblotting. The membranes were then incubated with primary antibody overnight at 4°C followed by the appropriate secondary antibody for 1 h at room temperature in PBS‐T/1% powdered milk. Following each step, the membranes were washed three times with PBS‐T for 10 min. Finally, the blots were developed using the enhanced chemiluminescence (ECL) system (NEN Life Sciences). The intensities of the signals were analysed by densitometry and normalized to the β‐actin signal using a digital imaging analysis system and 1D Image Analysis Software (Kodak Scientific Imaging Systems, USA). As a negative control, membranes were incubated with secondary antibody alone to validate the specificity of the signal.
The morphological characteristics of apoptosis were evaluated using Hoechst 33342 dye (Molecular Probes, USA), which stains the DNA of cells. Cells were incubated with 2 µg/ml of the Hoechst dye in Opti‐MEM for 15 min at room temperature and then visualized by fluorescent microscopy.
Following treatment, Jurkat cells (1 × 106) were collected and the adherent trophoblast cells (2 × 106) were detached with 0.05% trypsin–EDTA (Gibco) and both were centrifuged at 300 g for 10 min at 4°C. The pelleted cells were washed twice with 5 ml cold PBS. After the final centrifugation, the cellular pellet was resuspended in 1 ml cold PBS and incubated on ice with 5 µg/ml Hoechst 33342 dye (Molecular Probes) and 1 µg/ml propidium iodide (PI; Sigma) for 15 min. Hoechst 33342 dye stains the condensed chromatin of apoptotic cells more brightly than the chromatin of normal cells, whereas PI is only permeant to dead cells. The staining pattern that results from the simultaneous use of both these dyes makes it possible to distinguish between apoptotic and dead cells by flow cytometry. Unstained cells served as a measure of background fluorescence. The samples were analysed using a FACS Vantage (BD Biosciences) with 488 nm/UV dual excitation. PI staining was detected in the FL‐2 channel and Hoechst staining detected in the SSc‐W channel. Data were analysed using CellQuest software (BD Biosciences).
The data are represented as the mean ±SD and analysed for statistical significance (P < 0.05) using Student’s t‐test. All experiments were repeated three times with similar results.
First trimester trophoblast cells are resistant to Fas‐mediated apoptosis
Previous studies in our laboratory, as well as others’, have shown that first trimester trophoblast cells express the full‐length transmembrane form of Fas, but do not undergo apoptosis upon Fas stimulation under normal conditions (Payne et al., 1999; Aschkenazi et al., 2002). In order to confirm this observation, the viability of the first trimester human trophoblast cell lines, 3A and H8, as well as primary cultures were evaluated following treatment with an agonistic anti‐Fas mAb for 24 h using the CellTiter 96 assay. No difference in trophoblast cell viability could be detected between primary cultures (data not shown) and 3A cells treated with (99.6 ± 6.3%) or without (100 ± 2.2%) the anti‐Fas mAb (Figure 1A). Interestingly, the H8 cells exhibited a 61.7 ± 3.7% increase in cell viability upon Fas stimulation (P < 0.0005). In contrast, when the Jurkat T cell line, which is highly sensitive to Fas‐induced apoptosis, was treated with the anti‐Fas mAb, a 49.5 ± 4.3% decrease in cell viability was observed (P < 0.0005). Similarly, when the number of trophoblast cells undergoing apoptosis was quantified by flow cytometry, no difference could be detected between the untreated (20.9%) and anti‐Fas mAb‐treated 3A cells (9.4%) (Figure 1B). Conversely, the percentage of apoptotic Jurkat cells increased by 54.3% following Fas stimulation. This finding was confirmed at the molecular level, since only the pro‐forms of caspase‐8 (55 kDa), caspase‐9 (43 kDa) and caspase‐3 (33 kDa) could be detected in trophoblast cells treated with the anti‐Fas mAb for 24 h, whereas the Jurkat cells exhibited an increase in the activation of caspase‐8 (45/43 kDa), caspase‐9 (36 kDa) and caspase‐3 (17 kDa) following Fas stimulation (Figure 1C).
XIAP is highly expressed in first trimester primary trophoblast cells
Since the expression of Fas does not necessarily correlate with susceptibility to Fas‐induced apoptosis, intracellular inhibitors may exist downstream of Fas in order to prevent Fas‐mediated apoptosis in trophoblast cells (Payne et al., 1999; Aschkenazi et al., 2002). We hypothesized that XIAP may promote trophoblast survival by preventing the activation of the Fas apoptotic cascade in trophoblast cells. Therefore, the correlation between the expression and activation of XIAP and gestational age was evaluated in first trimester and term placental tissue samples. Densitometric analysis of XIAP expression levels revealed that XIAP was differentially expressed in first and third trimester placentas (Figure 2A). High levels of the active form of XIAP were observed in first trimester placentas (P < 0.05), but not in term placental tissue samples. In contrast, the predominant band detected in term placentas was the inactive fragment of XIAP (P < 0.05).
In order to determine if the trophoblast was the placental cell type responsible for the expression of XIAP, primary trophoblast cells were isolated from first trimester placentas between 6 and 12 weeks of gestation. As shown in Figure 2B, the first trimester primary trophoblast cells expressed only the active form of XIAP (45 kDa), whereas the inactive 30 kDa fragment of XIAP could not be detected. This suggests that XIAP may promote first trimester trophoblast cell resistance to apoptosis.
First trimester trophoblast cells undergo apoptosis upon XIAP inactivation
Our next objective was to determine whether the inactivation of XIAP caused first trimester trophoblast cells to undergo apoptosis. This was accomplished by treating the trophoblast cell lines and primary cultures with phenoxodiol, an inhibitor of XIAP (Kamsteeg et al., 2003). When the trophoblast cells were treated with 1 µg/ml of phenoxodiol, a time‐dependent increase in XIAP inactivation was observed, evidenced by the decrease in the expression of the full‐length p45 form and the increase in the expression of the p30 fragment of XIAP (Figure 3). Similar results were obtained with increasing doses of phenoxodiol for 24 h, with maximal XIAP inactivation at 10 µg/ml. As a potent caspase inhibitor, XIAP prevents the activation of caspase‐9 through its BIR3–RING domain and caspase‐3 activation via the linker–BIR2 domain (Takahashi et al., 1998; Sun et al., 2000). Consequently, when the 45 kDa form of XIAP was expressed, only the pro‐forms of caspase‐9 (43 kDa) and caspase‐3 (33 kDa) were present in H8 cells (Figures 3 and 1C). Following treatment with phenoxodiol, a decrease in pro‐caspase‐9 expression was observed and the cleavage products of caspase‐3 (17/19 kDa) were detected in both a time‐ and dose‐dependent manner. This increase in caspase activation correlated with the inactivation of XIAP, as demonstrated by an increase in the p30 form of XIAP (Figure 3).
In order to confirm that phenoxodiol‐induced XIAP inactivation resulted in trophoblast cell apoptosis, the trophoblast cell lines and primary cultures were treated with 10 µg/ml of phenoxodiol for 24 h and visualized by light microscopy. As shown in Figure 4A.2, trophoblast cells exhibited morphological characteristics of cells undergoing apoptosis following treatment with phenoxodiol. When the nuclei of the trophoblast cells were examined by fluorescent microscopy using Hoechst 33342 dye, DNA condensation was detected in the phenoxodiol‐treated 3A cells, a typical characteristic of apoptosis (Figure 4A.4). Moreover, a significant decrease in trophoblast cell viability (57.5 ± 1.6%) was observed in H8 cells following treatment with 10 µg/ml phenoxodiol for 24 h (P < 0.005; Figure 4B). This finding was further confirmed by flow cytometry using PI and the Hoechst 33342 dye. When trophoblast cells were treated with 10 µg/ml phenoxodiol for 24 h, the percentage of PI/Hoechst double positive H8 cells increased from 28.7% in the untreated control to 86.9% in the phenoxodiol‐treated cells (Figure 4C). Similar results were obtained from 6–12 week primary trophoblast cultures (data not shown). Therefore, the phenoxodiol‐induced increase in trophoblast cell death was due to the induction of apoptosis.
XIAP inactivation renders trophoblast cells sensitive to Fas‐mediated apoptosis
Our next objective was to determine whether XIAP inactivation rendered trophoblast cells sensitive to Fas‐mediated apoptosis. The trophoblast cell lines and primary cultures were pre‐treated with 1 µg/ml of phenoxodiol for 18 h, then treated for an additional 24 h with 1 µg/ml phenoxodiol alone or the combination of 1 µg/ml phenoxodiol and 500 ng/ml of the anti‐Fas mAb and visualized by light microscopy. As Figure 5A illustrates, trophoblast cells pre‐treated with phenoxodiol followed by the anti‐Fas mAb exhibited a further increase in H8 cell death in comparison with phenoxodiol alone. In order to determine if there was an additive effect between phenoxodiol and phenoxodiol plus anti‐Fas on cell viability, the viability of trophoblast cells following similar treatment was evaluated using the CellTiter 96 assay. Indeed, a significant decrease in trophoblast cell viability was observed in the phenoxodiol and anti‐Fas mAb‐treated 3A cells (38.6 ± 2.9%) compared with phenoxodiol alone (58.8 ± 4.3%) (P < 0.0005; Figure 5B). Similar results were obtained with primary trophoblast cultures (data not shown).
When the trophoblast cell lines and primary cultures were pre‐treated with 1 µg/ml of phenoxodiol for a period of 2 h, then treated with either 1 µg/ml phenoxodiol alone or the combination of phenoxodiol (1 µg/ml) and increasing concentrations of the anti‐Fas mAb for 24 h, an anti‐Fas dose‐dependent increase in XIAP inactivation (30 kDa) was observed only in the trophoblast cells pre‐treated with phenoxodiol followed by anti‐Fas treatment (Figure 5C). Upon XIAP inactivation, the trophoblast cells became sensitive to Fas stimulation, as exemplified by the increase in the active forms of caspase‐8 (45/43 kDa), caspase‐9 (36 kDa) and caspase‐3 (17/19 kDa) following increasing doses of the anti‐Fas mAb. Therefore, phenoxodiol‐induced XIAP inactivation sensitized trophoblast cells to Fas‐mediated apoptosis.
Apoptosis is important for the regulation of trophoblast survival and function during pregnancy. In the present study, we characterize the expression of XIAP in first trimester trophoblast cells and show that XIAP protects first trimester trophoblast cells from Fas‐mediated apoptosis. The Fas/FasL system is one of the most widely studied and best characterized apoptotic pathways. Fas and FasL were originally implicated in the clonal deletion of self‐reactive thymocytes in secondary lymphoid tissues, cytotoxic T cell killing and the removal of activated peripheral T cells following an immune response (Nagata, 1994; Van Parijs et al., 1998; Mor et al., 2001). Besides controlling the turnover of T lymphocytes, the Fas/FasL system is also thought to be involved in the establishment and maintenance of immune‐privilege sites, including the uterus (Uckan et al., 1997; Huppertz et al., 1998; Mor et al., 1998). Moreover, we and others have shown that Fas and FasL play an important role in mediating the tissue remodelling of reproductive tissues such as the mammary gland (Song et al., 2000), ovary (Sapi et al., 2002) and endometrium (Song et al., 2002). In addition, it has also been suggested that the Fas/FasL system may represent the mechanism by which the embryo breaches the epithelial barrier and adheres to the basement membrane of the maternal stroma during implantation (Galan et al., 2000). The precise role of the Fas apoptotic pathway in the regulation of trophoblast apoptosis, however, remains to be elucidated.
We initially demonstrated that first trimester trophoblast cells are resistant to Fas‐mediated apoptosis in spite of expressing both Fas and FasL. Consistent with previous findings (Payne et al., 1999; Aschkenazi et al., 2002), no difference in trophoblast cell viability could be detected between the untreated and anti‐Fas mAb‐treated 3A cells and primary trophoblast cultures. Interestingly, the H8 cells exhibited a statistically significant increase in cell viability upon Fas stimulation. Similarly, when the number of apoptotic trophoblast cells was quantified by flow cytometry, the percentage of trophoblast cell death actually decreased, which may be explained by the activation of survival pathways such as NFκB (Rensing‐Ehl et al., 1995). In addition, activation of the caspase cascade could not be detected in trophoblast cells following Fas stimulation, confirming that first trimester trophoblast cells are indeed resistant to Fas‐induced apoptosis.
Several mechanisms may explain the lack of a trophoblast cell response to Fas stimulation. At the receptor level, Fas‐mediated apoptosis can be inhibited by dominant‐negative decoy receptors, which lack death domains, receptor endocytosis or by transmembrane‐deficient soluble Fas (sFas) (Cascino et al., 1995). However, it has been previously shown that trophoblast cells predominantly express the full‐length transmembrane form of Fas (Payne et al., 1999; Aschkenazi et al., 2002) and trophoblast expression levels of Fas do not vary significantly throughout development (Gruslin et al., 2001), making this an unlikely explanation for trophoblast resistance to Fas‐induced apoptosis. Alternatively, propagation of the Fas death signal may be inhibited intracellularly at the level of DISC formation by FLIP, which precludes caspase‐8 recruitment to the DISC. Not only is FLIP capable of inhibiting Fas‐induced apoptosis, but it has also been shown to be constitutively expressed by the placenta (Irmler et al., 1997). In support of this, we recently demonstrated that interleukin‐10 (IL‐10)‐induced FLIP expression and activation protects first trimester trophoblast cells from Fas‐mediated apoptosis (Aschkenazi et al., 2002). As one of the predominant anti‐inflammatory cytokines produced at the maternal–fetal interface, IL‐10 promotes trophoblast survival by inhibiting the production of pro‐inflammatory cytokines, which are potentially harmful to pregnancy success (Casey et al., 1989; Yui et al., 1994; Raghupathy, 1997).
Bcl‐2 and Mcl‐1, members of the B cell lymphoma‐2 (Bcl‐2) family, have also been suggested to play a role in trophoblast survival (Marzioni et al., 1998). The Bcl‐2 family is comprised of both anti‐ and pro‐apoptotic members, which can differentially modulate death signals either directed towards, or initiated by, the mitochondrial pathway. By decreasing the permeability of the mitochondrial membrane to cytochrome c release (Adams and Cory, 1998; Gross et al., 1999), anti‐apoptotic Bcl‐2 and Mcl‐1 are capable of inhibiting the activation of caspase‐9 in trophoblast cells. However, Bcl‐2 and Mcl‐1 appear to be involved in preventing apoptosis during syncytial fusion since a higher expression is observed in syncytotrophoblast compared with cytotrophoblast cells (Huppertz et al., 1998; Marzioni et al., 1998). Moreover, there is also evidence suggesting that Bcl‐2 does not completely protect the syncytium from apoptosis (Smith et al., 1997b), which may depend on the stoichiometry of anti versus pro‐apoptotic Bcl‐2 family members in trophoblast cells.
In contrast to FLIP and anti‐apoptotic Bcl‐2 family members, IAP are unique in that they are capable of inhibiting both the mitochondrial and death receptor‐mediated pathways. Originally discovered in baculoviruses (Crook et al., 1993), IAP family members are characterized by varying numbers of baculoviral IAP repeat (BIR) domains, which are highly conserved from Drosophila to man (reviewed by Deveraux and Reed, 1999). X‐linked IAP (XIAP) consists of a C‐terminal RING finger domain and three tandem BIR domains, which have been shown to differentially inhibit both initiator and effector caspases (Deveraux et al., 1997). Besides inhibiting caspase function, XIAP has also been suggested to be involved in other cellular processes, including receptor‐mediated signal transduction and protein ubquitination via the RING finger domain (reviewed by Holcik et al., 2001). In addition, XIAP immunoreactivity was recently identified in the villous part of first trimester placentas, but not at term, suggesting an anti‐apoptotic role for XIAP in first trimester trophoblast cells (Gruslin et al., 2001). However, only the expression of the full‐length form of XIAP was characterized in placental tissue samples. In the present study, we demonstrate that the active form of XIAP was primarily expressed in first trimester placentas, whereas the predominant band detected in term placentas was the inactive form of XIAP. Although trophoblast cells are one of the most abundant placental cell types, stromal cells and immune cells such as macrophages may also be present in the placenta, which may contribute to the inactive form of XIAP observed in first trimester placental tissue samples. Moreover, trophoblast cells undergo constant cell turnover during implantation, and therefore trophoblast apoptosis does occur in the first trimester (Mochizuki et al., 1998; Chan et al., 1999; Danihel et al., 2002). Once these apoptotic trophoblast cells and/or contaminating cells were removed by primary trophoblast isolation, however, the inactive form of XIAP was no longer detectable. This suggests that XIAP plays a role in first trimester trophoblast resistance to Fas‐mediated apoptosis.
In order to determine if XIAP protects first trimester trophoblast cells from Fas‐induced apoptosis, XIAP was inactivated in the trophoblast cell lines and primary cultures with phenoxodiol, an inhibitor of XIAP (Kamsteeg et al., 2003). Indeed, phenoxodiol‐induced XIAP inactivation caused the trophoblast cells to undergo apoptosis, as evidenced by the decrease in trophoblast cell viability and activation of the caspase cascade in the phenoxodiol‐treated cells. More importantly, the inactivation of XIAP by phenoxodiol treatment rendered first trimester trophoblast cells sensitive to Fas stimulation. When the anti‐Fas mAb was added, an additional decrease in trophoblast cell viability was observed and this was statistically significant from the phenoxodiol‐only‐treated cells. This finding was confirmed at the molecular level by the anti‐Fas mAb dose‐dependent increase in caspase‐8 activation, which indicates that the Fas death receptor‐mediated pathway is functional. Interestingly, caspase‐8 activation was also observed in trophoblast cells treated with phenoxodiol alone and this may be due to previous interactions between death receptors and their ligands such as Fas and FasL, which may initiate the apoptotic cascade. However, the expression of XIAP may inhibit further activation of the pathway and once XIAP is inactivated by phenoxodiol treatment, the cascade is able to progress and the trophoblast cells undergo apoptosis. It may also be explained by a recent study demonstrating that once activated, ‘effector’ caspase‐3 is able to exert positive feedback and cleave ‘initiator’ caspase‐8, thereby further activating the apoptotic cascade (Engels et al., 2000). Therefore, phenoxodiol‐induced XIAP inactivation sensitized trophoblast cells to Fas stimulation, indicating that XIAP confers trophoblast resistance to Fas‐induced apoptosis.
In addition to Fas, there are several other death receptors expressed by trophoblast cells, including TNFRp55 and DR4/DR5, the death receptors for TNF‐α and TRAIL respectively (Yui et al., 1996; Phillips et al., 2001). Although TRAIL is not expressed by cytotrophoblast cells, a previous study demonstrated that trophoblast cells are resistant to recombinant TRAIL (Yui et al., 1994). Nevertheless, we and others have shown that trophoblast cells are sensitive to TNF‐α‐induced apoptosis (Yui et al., 1994; Aschkenazi et al., 2002). Since XIAP is common to both Fas and TNFRp55 death receptor‐mediated pathways, how trophoblast cells remain resistant to Fas stimulation, while sensitive to TNF‐α treatment, remains unclear. One explanation may be that TNF‐α acts downstream of XIAP or activates an alternative pathway (Wallach et al., 1999). Alternatively, the concentration of TNF‐α in the microenvironment may influence trophoblast survival (Baxter et al., 1999). Numerous studies in both humans and murine models indicate that anti‐inflammatory cytokines predominate over pro‐inflammatory cytokines including TNF‐α at the maternal–fetal interface during normal pregnancy (Casey et al., 1989; Yui et al., 1994; Raghupathy, 1997). In turn, anti‐inflammatory cytokines such as IL‐10 can up‐regulate anti‐apoptotic proteins (Aschkenazi et al., 2002), which may explain how the trophoblast is able to survive in vivo, but why trophoblast cells are sensitive to TNF‐α‐induced apoptosis in vitro.
Previous placental apoptosis studies have mainly focused on the correlation between the incidence of apoptosis and the stage of placental development or the trophoblast cell type expressing certain apoptosis‐related proteins (Smith et al., 1997b, 2000; Huppertz et al., 1998; Gruslin et al., 2001). In this study, we show that XIAP inactivation renders the normally resistant first trimester trophoblast cells sensitive to Fas‐mediated apoptosis. To our knowledge, this represents the first study demonstrating a functional role for the intracellular anti‐apoptotic protein, XIAP, in the regulation of the Fas apoptotic cascade in trophoblast cells. Since apoptosis is important for the normal physiology of pregnancy and a greater incidence of placental apoptosis is observed in pregnancies complicated by pre‐eclampsia and IUGR (Smith et al., 1997a; Allaire et al., 2000), our findings may provide insight into both normal placental development and placental dysfunction associated with abnormal pregnancy.
The authors would like to thank Dr Seth Guller for tissue collection and Thomas Taylor for his technical assistance with the flow cytometry experiments. This work was supported by a grant from the National Institutes of Health RO1 HD37137‐01A2 to G.M.